The present invention relates to novel ligands for asymmetric catalysis, as well as methods for making and using the novel compounds. The novel ligands, which are readily modifiable, contain both a phosphine and an amide functionality linked by a metallocene backbone. In addition, the novel compounds are phosphine-amides derived from a phosphine-amine, rather than a phosphine-carboxylic acid. The invention further relates to employing the resulting substantially enantiomerically pure ligands in a catalytically active composition comprising one or more phosphinometallocenylamide compounds in complex association with one or more Group VIb or Group VIII metals.
Asymmetric catalysis is the most efficient method for generating products with high enantiomeric purity, since the asymmetry of the catalyst is multiplied many times over in generating the chiral product. Such chiral products have found numerous applications as building blocks in, for example, single enantiomer pharmaceuticals and agrochemicals.
Asymmetric catalysts used to make such chiral products can be enzymatic or synthetic in nature. The synthetic catalysts have much greater promise than the enzymatic catalysts because of a much greater latitude in the types of reactions in which they may be used. Synthetic asymmetric catalysts usually contain a metal reaction center surrounded by an organic ligand. The ligand used is ordinarily of high enantiomeric purity, and is the agent inducing the asymmetry to the reaction product. Such ligands are, in general, difficult to make and therefore expensive.
As described by Richards, C. J. et al., Tetrahedron: Asymmetry 1998, 9, 2377-2407, asymmetric ferrocene derivatives have found great utility as ligands for asymmetric catalysis in reactions as varied as asymmetric hydrogenations, asymmetric Aldol reactions, asymmetric organometallic additions, and asymmetric hydrosilations. These ferrocene species usually are bidentate in nature, using a variety of ligating species. There are, however, no reported cases of metallocenyl ligands possessing one phosphine and one amide as the sole ligating groups.
Mixed phosphine-amides have recently been reported as ligands for asymmetric allylation reactions. See, Clayden, J. et al., J. Org. Chem. 2000, 65, 7033-7040; and Mino, T. et al., Tetrahedron:Asymmetry 2001, 12, 287-291. In addition, there have been reports of multi-dentate ligands that may, under certain conditions, function as phosphine-amide ligands, Trost, B. M. et al., Tetrahedron Lett. 1994, 35, 5817-5820; Butts, C. P. et al., J. Chem. Soc. Chem. Commun. 1999, 1707-1708; and Kim, Y. K. et al., J. Org. Chem. 2000, 65, 7807-7813. All of the ligands described in the foregoing have generally afforded only moderate enantioselectivities, and in no cases gave results above 90% ee (enantomeric excess). In addition, all of the previous phosphine-amide ligands are based on a 2-diphenylphosphinobenzoic acid amide; there are no cases of phosphine-amide ligands based on an amide of a phosphine-amine substructure, and none uses a metallocene backbone.
There continues to exist a need for stable, substantially enantiomerically pure ligands that enable one to make chiral products with which to create a variety of useful chemicals, such as pharmaceuticals and agrochemicals. cl BRIEF SUMMARY OF THE INVENTION
I have prepared novel, substantially enantiomerically pure phosphinometallocenylamides having the general structures below 
as bidentate ligands for asymmetric catalysis. The ligands are particularly useful in, for example, asymmetric allylation reactions, affording products with high enantiomeric excess.
In addition, described more fully below are both processes to make the novel bidentate ligands as well as processes that employ such ligands. Further, exemplary catalyst complexes incorporating the novel ligands are described.